A principle invoked for decades, and a factual premise that had not been checked.
Double effect
The principle that causing a harm as a foreseen side effect can be permissible where causing it as a means or an end would not.
- Theme
- Ethics and character
- Tradition
- Medieval
- Term fixed
- c. 1270
In plain terms
There is a difference between doing something in order to bring about a harm and doing something knowing a harm will follow.
The bomber who targets a factory and foresees that civilians nearby will die is doing something different from the bomber who targets the civilians in order to break morale - even if the number of deaths is identical and both were certain.
Whether the difference matters morally is genuinely disputed. Consequentialists say no: the deaths are the same and the intention is a fact about the agent's psychology rather than about what happened. Defenders say the distinction is one everyone actually uses, that it is why murder and manslaughter are different offences, and that a theory unable to accommodate it has lost track of something ordinary.
As a philosopher would put it
The standard formulation sets four conditions: the act itself must not be wrong; the bad effect must not be the means to the good effect; the bad effect must be foreseen but not intended; and there must be proportionate reason for permitting it.
The second condition does most of the work and is the hardest to apply. Blowing up a person blocking the exit of a cave uses their death as the means and is excluded; a bombing raid that kills the same number as a side effect is not - a distinction the closeness problem attacks, since an effect can be so tightly bound to the intended one that calling it unintended looks like a description of the agent's self-image rather than of the act.
Foot's trolley discussion introduced the case that made the whole area a research programme, and her own view was that the distinction between doing and allowing does more work than double effect. In just war theory the doctrine underwrites the principle of discrimination, and in medicine it governs pain relief at the end of life.
Where you meet it
Palliative sedation is used when symptoms cannot be controlled by other means: a patient close to death is sedated, sometimes to unconsciousness, to relieve refractory distress.
The doctrine of double effect is the standard justification, and it is cited by name in professional guidance in several countries. The reasoning is that the intention is to relieve suffering, that any shortening of life is foreseen and not intended, that the sedation is not a means to death but to comfort, and that the reason is proportionate given the circumstances.
The empirical premise has been examined and is weaker than the argument assumes: several studies find that properly titrated sedation does not measurably shorten survival, which removes the bad effect the doctrine was invoked to license. That is an unusual and instructive outcome. A principle developed to permit an act turned out to be needed less than expected, because the factual claim underneath it had not been checked for most of the period it was used.
What is actually disputed
Whether intention affects permissibility.
The case against is that an act's rightness should depend on what it does, not on how the agent describes their own aim. Thomson and others press the point that the doctrine appears to let an agent's self-understanding determine what is permitted, which gets the direction of explanation backwards - intentions bear on blameworthiness rather than on whether the act may be done.
Defenders reply that intention is not merely psychological: what a person intends determines what act they are performing, and using someone's death as a means is a different act from causing it incidentally, even where the physical movements coincide.
The closeness problem remains unresolved. If an effect is inseparable from the intended one, the claim that it was not intended has no purchase - and there is no agreed criterion for when two effects are close enough to collapse.
Commonly confused with
- the ends justify the means The doctrine says nearly the opposite: a bad effect used as a means is exactly what it forbids.
- unintended consequences Those are effects nobody foresaw. The doctrine is about effects that are foreseen with certainty and still not intended.
- collateral damage A military euphemism whose justification appeals to the doctrine. Using the phrase as if it were the principle skips the argument.
- plausible deniability The doctrine requires the effect to be genuinely unintended, and the closeness problem exists precisely because that is hard to establish.
Defined against
- Consequences If only outcomes determine rightness, then a distinction resting entirely on the agent's aim can make no difference
Philosophers
- Thomas Aquinas The self-defence discussion the doctrine grows from
- Immanuel Kant The means-end distinction the doctrine relies on
- Peter Singer The consequentialist rejection of the distinction
- Al-Farabi Just conduct in war, in the Islamic philosophical tradition
Works
- Summa Theologiae Where the germ of the doctrine appears
- Groundwork of the Metaphysics of Morals The formula of humanity, which the second condition echoes
- Intention The analysis of intention the modern debate depends on
- An Introduction to the Principles of Morals and Legislation The utilitarian framework it constrains
- After Virtue The historical setting of the moral vocabulary
Schools
- Thomism Where the doctrine is developed
- Natural Law Where it is standardly deployed
- Classical Utilitarianism Where it is rejected
- Cambridge Analysis Where the modern analysis is conducted
- Neo-Thomism Where it is defended in the twentieth century
Related concepts
Consequences 1789 · Duty 1785 · Responsibility c. 340 BC · Moral luck 1976
Aquinas's discussion of self-defence contains the germ; the four-condition formulation is later Catholic moral theology, and the doctrine's modern career is largely medical and military.